That's not a blue moon; pull over...

March 1, 2012 2:47 PM

Can anyone point me to actual data on how often motorists receive traffic tickets? I'm looking for where the top of the bell curve is, so to speak - obviously, individuals vary widely, which is kind of the point of my question - how often does the "median" driver receive traffic tickets?

Google is failing me - I can find data like how many citations are issued, and found one stat that said "1 in 6 motorists receive a citation annually," but that doesn't really tell me what I want to know, as all motorists do not have an equal chance of receiving one (and some get more than one).

Anecdotes welcome, but probably won't really help unless they involve a database. Of survey data or insurance data about average number of citations. :-)

posted by randomkeystrike to law & government (12 answers total)

(ms. Vegeabtle)
I don't have time to search for specifics right now, but have you tried iihs - the insurance institute for highway safety? This seems like something they should have data on.
posted by a robot made out of meat at 2:58 PM on March 1, 2012 [1 favorite]

Thanks for the idea - haven't found anything there.
posted by randomkeystrike at 3:38 PM on March 1, 2012

I have no database, but plenty of speculation, to give you.

My gut is that there are an enormous number of people who've never been pulled over at all, and the folks at the top of the bell curve will have one ticket in a driving lifetime of 40 years. I was thirty when I was pulled over for the first time, my mom was in her forties, and my dad was in his forties. I think the combined lifetime total of actual citations for all adult drivers in my immediate family (five parents, four adult children) is 9 or 10, with a median time-as-a-driver at something like 22 years. Of the nine of us, three have never received a ticket (including me; I just got a warning.) The middle three in terms of number of tickets received each had one ticket - the next had two, and then the top two people have had 3 tickets each.

Note this chart and the average age of cited drivers. I think most people just aren't that unlucky/stupid to get more than a few tickets (again, most people) and that the average there is the age of a second or third citation at most. That'll put you well into your sixties, for most of those cars, by the time you get your fourth.

(The percentages are useless - they're comparing the average number of citations per type car to the number of citations for that type of car; there are no hard numbers)
posted by SMPA at 3:45 PM on March 1, 2012

The percentages are useless

They do give one bit of insight relevant to the question--the lowest vehicle had only 23% of the "average" number of citations, while the highest vehicle had 404% of average.

Or in other words, the drivers of one vehicle received about 18 times the number of tickets of drivers of the other vehicle.

So that doesn't tell us what the median is, but it certainly does confirm the anecdotal point you made--that there is a huge amount of variation in the number of tickets that different types of drivers receive.
posted by flug at 5:31 PM on March 1, 2012

I saw that chart - I think all it does is give you an idea of how often the CAR gets cited above the mean. A car with 100% in the column instead of 400% gets pulled over an exactly average rate - but those numbers tell you nothing about what that rate is (not that that would answer my question, exactly). The average age of the driver doesn't tell you anything about how often the drivers of those cars have been cited. Presumably, since the drivers control the car, and not the other way around, drivers of that "top" rated Mercedes model have gotten tickets before they were in their 50s.

And yeah, it's without question that the bell curve is pretty long from one end to another. I had a friend in high school who got so many citations that he lost his license at one point - he'd only been licensed less than two years and written up 6-8+ times. I think the statistical term for this guy is "idiot."
posted by randomkeystrike at 5:40 PM on March 1, 2012

I think most people just aren't that unlucky/stupid to get more than a few tickets (again, most people) and that the average there is the age of a second or third citation at most.

There are a lot variables that this leaves out. Location is important as are the actual roads traveled. Also, for some (cough cough) it has nothing to do with luck or stupidity. Some make a calculated risk reward of time v cost of getting caught. I have never had an accident in 30+ years of driving, but I average a speeding ticket every 2 years for the last 10, but had only 1 for the years before that. I do a lot of highway driving now.

Also, if you include EVERY licensed driver, you will be pulling in folks like my friend who lived in NYC for 20 years and never drove once during that time and older folks who drive rarely as well as commercial drivers. Also, there are times when you get pulled over and cited with multiple moving violations. Also, are you looking for convictions or citations? I have been cited way more than convicted. I have plead guilty to lesser charges too. For example, rather than accept a speeding ticket I plead guilty to a charge of ignoring a highway sign. No points, but more money. That sign? The speed limit sign, natch.

I think the best source would be insurance companies or the insurance institute (call them and ask in addition to checking their site). I would also look at a government agency such as the institute for highway safety or whomever does the crash test dummy thing for the government. I think you will be more likely to find the data on a state by state basis.

My random guess would be that the median driver gets a speeding ticket or moving violation once every 12 years.
posted by AugustWest at 8:42 PM on March 1, 2012 [1 favorite]

Hm. I can't seem to find what you're interested in, which is too bad, because I'm interested in it, too. One data point: this paywalled, self-reported survey says first-time drivers have a 15% chance of getting a ticket their first year of driving. Actually, that's not that big of a risk, in the scheme of things, just 1 in 6. I would have thought it was higher, especially given that the same survey says the risk of a crash is 1 in 3, for the first year.

One small study of 100 cars observed drivers and recorded "crashes, near-crashes, and incidents." It is, as you would expect, a classic "hockey stick" pattern.

Florida, in 1998, had 800K citations for 13M drivers, 1 in 16 drivers. Hope you find the data you're looking for.
posted by wnissen at 8:43 PM on March 1, 2012 [1 favorite]

wnissen - that NHTSA study, although pretty small, as you said, does have data in figure 4.6 that almost answers my question, if I'm reading it right - one class of drivers typically had an average of slightly more than 1 in 5 years; another class (the more "involved", to use the survey's terms?) has 2.-something in 5 years on average.

1 in 16 also roughly correlates to - a 31% chance per five years? Am I working my additive property of probability right? Of course, the problem is that hockey stick pattern, again. Some drivers get bunches of them, others never, and something tells me the average is not close to the median in this case.
posted by randomkeystrike at 7:41 AM on March 2, 2012

In the interest of my locale's tourism industry, I will spare you all the details of my driving ineptitude. Though I am, I believe, five years clean. I'm sure there are others like me.

The more you know.
posted by teekat at 7:47 AM on March 2, 2012

Ooh, it appears I am anonymous. Well, I still won't tell you.
posted by teekat at 7:50 AM on March 2, 2012

Not a direct answer but still possibly helpful: the distribution probably is not a bell curve (normal) distribution. It is a count variable, which means that many people will have few tickets per [timeframe], and few people will have many. So the distribution will have a loooooong right tail and a relatively truncated lefttail. (like the functions for lambda 1-3 here: http://paulbourke.net/miscellaneous/functions/poisson4.gif)

Bearing all this in mind, median's an okay measure of central tendency, but I'd prefer the mode.
posted by powerbumpkin at 10:54 PM on March 2, 2012

randomkeystrike, I see the table you're talking about. Sort of a weird way to divide things up, by the average number of events, but it also brings up a difficulty in collecting the data: when trying to gather information about long periods, self-reporting is really unreliable. For instance, do you really believe that the worst 25% of drivers (the "high involvement" group) has the same average number of collisions as the best 75%?

To really gather this data, you'd need to survey the same people over time, or else use a DMV database.
posted by wnissen at 3:34 PM on March 3, 2012